Does cognitive-behavioral therapy response in youth with obsessive-compulsive disorder differ if treatment ends during summer?

Among depressed adolescents, selective serotonin reuptake inhibitor (SSRI) and cognitive behavioral therapy (CBT) response differs as a function of the time during which the intervention is concluded, with improved outcomes associated with completing treatment during the summer (Shamseddeen, et al., 2011). This finding suggests that the timing of treatment termination may be relevant to consider; yet, outside of depressed adolescents, no data have been reported in other pediatric psychiatric conditions. Accordingly, we examined timing of treatment termination on CBT outcome in youth with obsessive-compulsive disorder. One hundred and ninety-seven youth (113 males) with a principal diagnosis of OCD and a mean age of 12.54 years (SD = 2.91) received14 weekly or daily CBT sessions. Thirty-two participants had a comorbid depression diagnosis while 139 were taking psychiatric medication. Twenty-four youth were taking psychiatric medication and also had a comorbid depression diagnosis. Prior to treatment initiation, all participants had a score ≥16 on the Children’s Yale Brown Obsessive-Compulsive Scale (CY-BOCS; Scahill et al., 1997).Obsessive-compulsive disorder symptom severity and responder status were assessed via the Clinical Global Impression-Severity and -Improvement Scales (CGI-S,CGI-I; Guy, 1976). Participants with a score of “much” or “very much improved” on the CGI-I were considered treatment responders. Following Shamseddeen et al. (2011), those who concluded treatment between September and May were categorized as completing treatment during the school year (n=123), while the remainder of the sample were categorized as terminating in the summer time (June-August; n=74). Pre-treatment comparisons via t-tests and analysis of variance did not detect differences at the p < .05 level for the CY-BOCS and CGI-S scores between those who terminated in the summer time and those who terminated during the school year. Analysis of covariance (where group assignment predicted post-treatment scores while including pre-treatment scores as a covariate for each outcome of interest)did not demonstrate significant group differences on post-treatment CY-BOCS (Summer-M=10.47, SD=7.02; School-M=10.89, SD=7.48; F(1,194)=.09, p=.76, d=.06) and CGI-S ratings (Summer-M= 1.74, SD=1.05; School-M=1.78, SD=1.20; F(1,194)=.03, p=.87, d=.04).Logistic regression determined that group assignment did not significantly predict responder status (p=.51, Nagelkerke R2=.003). In sum, we found that in a large clinical sample, CBT for pediatric OCD demonstrated comparable outcomes regardless of when treatment is concluded. Unlike depression which may be more impacted than OCD by external stressors common to the school year (Shamseddeen et al., 2011), obsessive-compulsive symptoms may be more immune to external stressors, with improvement reflecting actual changes in the underlying pathology versus improvements in psychosocial stressors that may occur during the summertime (e.g., reduced school stress). Limitations include that less impaired cases could perhaps delay treatment until summer to avoid missing school. An additional weakness is that we lack data on whether youth were enrolled in school and whether schools followed year-round vs. traditional 9 month calendars. Also, a predominantly male, younger sample differs from that of Shamseddeen et al. (2011) which may contribute to the dissimilar findings. These sample characteristics could also limit generalizability. Ultimately, these data suggest that the timing of CBT termination does not confound outcomes in pediatric OCD and thus, does not need to be considered in analytic plans. However, the current study examined the effect of termination timing on outcomes based on the school year. Empirical findings suggest differences in platelet serotonin between seasons in pediatric OCD relative to controls (Brewerton, Flament, Rapoport, & Murphy, 1993) and a general pattern of symptom reduction in OCD symptoms falling between August and October (Kovalenko, et al., 2000). Future research should consider the effect of seasonality on outcomes, as well as, the role of termination timing in psychopharmacology treatment response in this population.